Hello everybody!We are new in Mikrotik devices,we have got a 1 km backhaul link:
Point A:24 dBi 5 Ghz Grid,RB 411 GL with R52nM working in AP Bridge with Nv2
Point B:27 dBi 5 Ghz Grid,RB 411 GL with R52nM working in Station Bridge
CCQ:80-100%,TX/RX Signal Strenght:55-55 dBm,Signal to Noise:60 dB
Our result is only 14-15 Mbit/sec TCP speed
I want to insert here picture but i don’t know how to
Thank you for your help
80-110mbps
So,this is sure interference problem,we are planning later 20km link with these devices,what do you think?What is the expectable bandwith from this kit?
We saw issues on a very short hop with a pair of RBSXT’s if I recall correctly the way it was we got 80~90 Mbps UDP traffic however when it came to TCP traffic it topped out at 25 ~ 30 Mbps spent a while looking at this If I recall correctly it was down to queue size/type on the interface. If you need a work around for this try and pipe your traffic down an EOIP tunnel as this encap’s all traffic in a udp header. don’t forget to allow the MTU’s out to allow for the extra overhead
hope this might help even if the UDP tunnel is between the wireless interfaces.
Hope this helps
you could quickly try a pair of MT Sextants or SXT’s in 5ghz to rule out hardware or cable / antenna / power supply issues with your current setup.
they are very inexpensive 99$ a pc OR LESS and you can just use them for testing when you are done (or use them for the link)
Thank you very much for your answers,we are beginner in Mikrotik devices,but this link is properly for learning.
At this moment we don’t want to buy another devices,maybe later.We assume the Mikrotik’s Bandwith measure is not correct
Can u post nv2 UDP test
and test with nstreme
Do u use only one polarization
1km 24-27db antenna is too much - I m not saying its not ok but why
Have u ever try doing this setup with SXT - it is going to be much much cheaper and better and smaller
Just to share this I have also bad experience with R52nM I m not going in to detail how and what, this card have same chipset as R52Hn exept Hn is 2db stronger but outside it is inferior to nM - dont know why
Also my Distributer confirm this - He have same complaints from some clients - but maybe I m wrong.
How can i insert here picture?Sorry for my bad english
First Browse your file than add the file
and at the end place inline
The place inline is does not appear for me
First browse and add the file
Than it will appear
Ok,here is the picture,the result is same with 20 TCP count too
finally Trackboy picture 
Signal to Noise is a bit high - U can change some freq
but it seam fine u use only one polarization u should get something around 40-50 mbits
I m not sure if 411 can generate so much traffic
and when u do testing add TCP connection count 10 or 20 (because nv2)
Can u do UDP test also test it with nstreme
And u dont have to add picture just post the results.
Why tcp count 1? Set it to 20 as it is default.
Any RB 4xx or 7xx with 400mhz can generate ~30mbps and the ones at 680mhz can generate ~100mbps. All TCP speeds.
Only my RB800 can generate 100mbps+
110mbps my RB450G 
Ok,i will test nstreme with UDP,with 20 TCP count the speedtest result is same
did you try a file transfer, torrent or anything else? Maybe the only problem is btest…
Ok,seems to be the 80 meters utp cable on ap side between RB2011 and RB 411 GL,not ok,TCP speed only 13-14 Mbit sec,we have to change
So the 411 GL capable of 100 Mbit/sec TCP speed?(50/50 Mbit/sec?)
But with the new Mikrotik mini PCI card the experience is not something good?
Try a UDP test before replacing cable.